| *** smoriya has joined #openstack-ironic | 00:00 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Use oslo.utils and oslo.concurrency https://review.openstack.org/138116 | 00:07 |
|---|---|---|
| * NobodyCam steps away to look for some food | 00:07 | |
| *** arif-ali has quit IRC | 00:09 | |
| *** arif-ali has joined #openstack-ironic | 00:12 | |
| JayF | https://review.openstack.org/#/c/92627/13 needs a final core review to land | 00:13 |
| *** romcheg1 has joined #openstack-ironic | 00:19 | |
| jroll | ^ +A | 00:21 |
| *** romcheg has quit IRC | 00:21 | |
| NobodyCam | lol /me stops looking | 00:21 |
| jroll | NobodyCam: actually | 00:21 |
| jroll | my name is on that patch | 00:21 |
| JayF | NobodyCam: it's okay, jroll was the original author, you should probably look | 00:22 |
| jroll | so your review would be good | 00:22 |
| jroll | but I don't think I wrote any of that code :P | 00:22 |
| JayF | NobodyCam: also it won't merge cleanly so someone has to rebase it | 00:22 |
| *** romcheg has joined #openstack-ironic | 00:31 | |
| NobodyCam | any reason you would want to add a log around line 78 of :https://review.openstack.org/#/c/92627/13/ironic_python_agent/netutils.py | 00:33 |
| *** romcheg1 has quit IRC | 00:33 | |
| JayF | Seems like a useful addition, and another patchset has to be published now anyway, so ... | 00:34 |
| NobodyCam | thats about all I see | 00:35 |
| *** Masahiro has joined #openstack-ironic | 00:38 | |
| *** eghobo has joined #openstack-ironic | 00:42 | |
| *** Masahiro has quit IRC | 00:43 | |
| *** andreykurilin_ has quit IRC | 00:44 | |
| *** romcheg has left #openstack-ironic | 00:44 | |
| *** eghobo has quit IRC | 00:45 | |
| *** Marga_ has quit IRC | 00:47 | |
| *** eghobo has joined #openstack-ironic | 00:47 | |
| *** eghobo has quit IRC | 00:49 | |
| *** Marga_ has joined #openstack-ironic | 00:52 | |
| *** eghobo has joined #openstack-ironic | 00:52 | |
| *** smoriya has quit IRC | 00:55 | |
| *** Marga_ has quit IRC | 00:56 | |
| *** eghobo has quit IRC | 01:02 | |
| *** eghobo has joined #openstack-ironic | 01:03 | |
| *** eghobo has quit IRC | 01:04 | |
| *** smoriya has joined #openstack-ironic | 01:06 | |
| *** spandhe has joined #openstack-ironic | 01:24 | |
| *** killer_prince is now known as lazy_prince | 01:26 | |
| *** shakamunyi has quit IRC | 01:28 | |
| openstackgerrit | Merged openstack/ironic: Fix image_info passed to IPA for image download https://review.openstack.org/139988 | 01:35 |
| *** marcoemorais has quit IRC | 01:40 | |
| *** marcoemorais has joined #openstack-ironic | 01:41 | |
| *** jerryz has joined #openstack-ironic | 01:42 | |
| *** alexpilotti has joined #openstack-ironic | 01:44 | |
| *** ryanpetrello has joined #openstack-ironic | 01:45 | |
| *** chenglch has joined #openstack-ironic | 01:48 | |
| *** ChuckC has quit IRC | 01:50 | |
| *** alexpilotti has quit IRC | 02:06 | |
| adam_g | anyone happen to know if its possible to push MTU settings per neutron port via extra_dhcp_opts? | 02:07 |
| *** ryanpetrello has quit IRC | 02:10 | |
| Haomeng | adam_g: not sure if veth_mtu in ovs_neutron_plugin.ini can help you | 02:11 |
| jroll | adam_g: I kind of doubt it, but I'd ask neutron | 02:12 |
| *** ChuckC has joined #openstack-ironic | 02:12 | |
| cuihaozhi | Haomeng: morning :) | 02:16 |
| *** ChuckC has quit IRC | 02:16 | |
| *** ChuckC has joined #openstack-ironic | 02:16 | |
| adam_g | Haomeng, was hoping to set it on the client side. you can set 'dhcp-option-force=26,1300' in a file and configure the dhcp agent to pass as an additional --conf-file to dnsmasq, but curious if thats possible in the same way we set up tftp/pxe stuff per-port | 02:17 |
| *** lazy_prince has quit IRC | 02:20 | |
| *** Masahiro has joined #openstack-ironic | 02:26 | |
| *** marcoemorais has quit IRC | 02:31 | |
| *** Masahiro has quit IRC | 02:31 | |
| *** marcoemorais has joined #openstack-ironic | 02:31 | |
| *** killer_prince has joined #openstack-ironic | 02:33 | |
| *** killer_prince is now known as lazy_prince | 02:33 | |
| *** rloo has quit IRC | 02:40 | |
| Haomeng | adam_g: ok | 02:41 |
| *** ramineni has joined #openstack-ironic | 02:43 | |
| ramineni | JoshNang: hi | 02:44 |
| *** marcoemorais has quit IRC | 02:48 | |
| *** dlaube has quit IRC | 03:11 | |
| *** spandhe has quit IRC | 03:20 | |
| *** david-lyle is now known as david-lyle_afk | 03:24 | |
| *** ryanpetrello has joined #openstack-ironic | 03:27 | |
| *** Masahiro has joined #openstack-ironic | 03:42 | |
| ramineni | jroll, JayF: hi | 03:44 |
| *** Masahiro has quit IRC | 03:47 | |
| *** bradjones has quit IRC | 03:51 | |
| *** achanda has joined #openstack-ironic | 03:59 | |
| *** ryanpetrello has quit IRC | 04:02 | |
| *** dlaube has joined #openstack-ironic | 04:06 | |
| *** achanda has quit IRC | 04:17 | |
| *** achanda has joined #openstack-ironic | 04:17 | |
| *** rameshg87 has joined #openstack-ironic | 04:21 | |
| *** lazy_prince has quit IRC | 04:22 | |
| *** achanda has quit IRC | 04:22 | |
| *** lazy_prince has joined #openstack-ironic | 04:22 | |
| *** wanghua has quit IRC | 04:24 | |
| *** pcrews has quit IRC | 04:29 | |
| *** chenglch has quit IRC | 04:33 | |
| *** rushiagr_away is now known as rushiagr | 04:34 | |
| *** achanda has joined #openstack-ironic | 04:37 | |
| *** achanda has quit IRC | 04:40 | |
| *** Masahiro has joined #openstack-ironic | 04:43 | |
| *** Masahiro has quit IRC | 04:47 | |
| devananda | evening, all | 04:51 |
| devananda | reminder -- meeting is in 10 minutes | 04:51 |
| * devananda wonders if anyone is around :) | 04:55 | |
| lintan | I will join | 04:55 |
| jroll | ramineni: hi, what's up? | 04:55 |
| ramineni | jroll : hi , IPA supports disk_erase during zapping right? which tool IPA uses to do disk_erase? | 04:57 |
| jroll | ramineni: not upstream (yet) but yes | 04:58 |
| jroll | it uses hdparm for normal SSDs | 04:58 |
| ramineni | jroll : is it opensource tool? I want to explore the same for our servers too :) | 04:58 |
| mrda | \o | 04:59 |
| victor_lowther | hdparm has been opensource for at least 15 years. :) | 04:59 |
| jroll | ramineni: https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/hardware.py#L272-279 | 04:59 |
| jroll | ramineni: ironic does not call that command upstream at the moment, though | 04:59 |
| * victor_lowther remembers using it to enable IRQ batching and MWDMA on old ATA controllers for great performance boosts. | 05:00 | |
| jroll | ramineni: this is how we do our fancy LSI nytro cards https://github.com/rackerlabs/onmetal-ironic-hardware-manager/blob/master/onmetal_ironic_hardware_manager/__init__.py#L181 | 05:00 |
| ramineni | jroll : thanks , will check :) | 05:00 |
| jroll | :) | 05:00 |
| devananda | meeting time! | 05:00 |
| devananda | remember, new room -- #openstack-meeting-3 | 05:00 |
| jroll | ramineni: we're running it in production today, it works well | 05:00 |
| jroll | ramineni: however, we don't erase spinning disks at the moment | 05:01 |
| Haomeng | ok | 05:01 |
| ramineni | victor_lowther: :P , my bad , havent used it..so no idea | 05:01 |
| ramineni | jroll: oohok | 05:01 |
| * jroll wonders what time it is for dtantsur|afk and friends | 05:03 | |
| sambetts | jroll: Its 5am for me and 6am for dtantsur|afk | 05:07 |
| jroll | aha | 05:08 |
| jroll | thanks :) | 05:08 |
| *** rushiagr is now known as rushiagr_away | 05:09 | |
| sambetts | jroll: what time is it for you guys? | 05:10 |
| jroll | sambetts: 9pm in cali | 05:11 |
| sambetts | ah, thanks | 05:12 |
| sambetts | :) | 05:13 |
| *** pensu has joined #openstack-ironic | 05:14 | |
| *** harlowja is now known as harlowja_away | 05:16 | |
| *** dlaube has quit IRC | 05:17 | |
| devananda | NobodyCam: dont suppose you're around? | 05:24 |
| *** dlaube has joined #openstack-ironic | 05:29 | |
| *** dlaube has quit IRC | 05:30 | |
| *** Nisha has joined #openstack-ironic | 05:30 | |
| *** yuanying_ has quit IRC | 05:37 | |
| *** Masahiro has joined #openstack-ironic | 05:44 | |
| *** rakesh_hs has joined #openstack-ironic | 05:45 | |
| *** yuanying has joined #openstack-ironic | 05:46 | |
| *** Masahiro has quit IRC | 05:48 | |
| *** faizan_ has joined #openstack-ironic | 06:02 | |
| jroll | lintan: hmm, maybe you're right, idk | 06:02 |
| jroll | like | 06:03 |
| jroll | this has been tested in devstack | 06:03 |
| jroll | with dnsmasq | 06:03 |
| *** chenglch has joined #openstack-ironic | 06:03 | |
| lintan | jroll: so it's wired | 06:04 |
| jroll | hm? | 06:05 |
| lintan | jroll: I didn't try devstack but it doesn't work | 06:05 |
| jroll | interesting | 06:05 |
| sirushti | Hi all, I brought this up in the meeting but could this spec https://review.openstack.org/#/c/97150/ get some reviews please? | 06:06 |
| jroll | lintan: sorry, but I need to go, let's chat more later | 06:06 |
| jroll | g'night everyone | 06:07 |
| lintan | jroll: OK, lucas test on my commit and approve it, we can ask him next week | 06:07 |
| jroll | ok, thanks | 06:07 |
| lintan | good night | 06:07 |
| devananda | I'm also fading fast ... see everyone tomorrow! | 06:10 |
| *** lazy_prince has quit IRC | 06:13 | |
| *** rushiagr_away is now known as rushiagr | 06:14 | |
| mrda | night devananda | 06:14 |
| *** Marga_ has joined #openstack-ironic | 06:14 | |
| Haomeng | devananda: night:) | 06:21 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic-specs: fix the filename for hardware-capabilities.rst https://review.openstack.org/140252 | 06:22 |
| *** killer_prince has joined #openstack-ironic | 06:25 | |
| *** killer_prince is now known as lazy_prince | 06:25 | |
| *** faizan_ has quit IRC | 06:39 | |
| *** rameshg87 has quit IRC | 06:41 | |
| *** rameshg87 has joined #openstack-ironic | 06:41 | |
| jiangfei | hi, haomeng | 06:42 |
| Haomeng | hi | 06:42 |
| Haomeng | jiangfei: hi | 06:42 |
| *** russellb has quit IRC | 06:43 | |
| *** lazy_prince has quit IRC | 06:44 | |
| jiangfei | node-create should valid the input driver do you mean, we use curl -x post /node -d '{"driver": "xxxxxx"}' has faultstring .but ironic node-create just Bad Request (HTTP 400), no faulutstring | 06:44 |
| *** russellb has joined #openstack-ironic | 06:44 | |
| *** lazy_prince has joined #openstack-ironic | 06:45 | |
| Haomeng | jiangfei: yes | 06:45 |
| Haomeng | jiangfei: it is better that we valid the driver, and tell user the input driver is invalid | 06:46 |
| Haomeng | jiangfei: you can run with command line | 06:46 |
| jiangfei | Haomeng: it's also modify the error message? | 06:46 |
| Haomeng | jiangfei: I am prepare to commit the patch, but not sure if it is handled by cliend side | 06:46 |
| Haomeng | jiangfei: two cases, the one is that the driver is invalid/not enabled, it is not listed in our ironic.conf enable_driver | 06:47 |
| jiangfei | Haomeng: yes | 06:47 |
| Haomeng | jiangfei: another one is that the driver is valid, it is listed in conf, but no any conductor support such driver | 06:47 |
| Haomeng | jiangfei: so can not find the conductor | 06:48 |
| jiangfei | Haomeng: nice | 06:48 |
| Haomeng | jiangfei: maybe no any conductor is started | 06:48 |
| Haomeng | jiangfei: but it is complex if we have multi conductor | 06:48 |
| Haomeng | jiangfei: so current behavior is that we try to lookup all conductors to find which supports the input driver | 06:49 |
| jiangfei | Haomeng: yes | 06:49 |
| Haomeng | jiangfei: in fact, our drivers are plugable, so we can check from python exts to see if it is loaded already | 06:49 |
| Haomeng | jiangfei: any idea here? | 06:50 |
| jiangfei | Haomeng: i would like to ' check from python exts to see if it is loaded already' | 06:51 |
| Haomeng | jiangfei: yes, maybe the conductor side code to be changed to check python exts first | 06:52 |
| Haomeng | jiangfei: then to check for each conductors | 06:52 |
| Haomeng | jiangfei: but have to check each conductor env's python exts | 06:52 |
| Haomeng | jiangfei: the conductors can be distributed | 06:52 |
| Haomeng | jiangfei: so here some complex | 06:53 |
| jiangfei | Haomeng: the conductors can has different drivers? | 06:53 |
| Haomeng | jiangfei: yes, it can be installed on different nodes with different confs | 06:53 |
| *** marcoemorais has joined #openstack-ironic | 06:56 | |
| *** marcoemorais1 has joined #openstack-ironic | 06:58 | |
| jiangfei | Haomeng: may be we should add retry times in conductors in this case. | 07:00 |
| *** marcoemorais has quit IRC | 07:01 | |
| Haomeng | jiangfei: I think we just check from conductor side to see if python exts is loaded | 07:01 |
| jiangfei | Haomeng: if not return the error message , or try to check it the other conductors | 07:03 |
| *** Marga_ has quit IRC | 07:05 | |
| Haomeng | jiangfei: I am not sure if driver-list can list all conductors drivers | 07:05 |
| Haomeng | jiangfei: if it is case, we can check with driver-list result | 07:05 |
| Haomeng | jiangfei: have to check driver-list code logic | 07:06 |
| jiangfei | Haomeng: yes | 07:06 |
| Haomeng | jiangfei: :) | 07:07 |
| Haomeng | jiangfei: if you'd like to fix this, go ahead please | 07:07 |
| Haomeng | jiangfei: dont worry | 07:07 |
| jiangfei | Haomeng: thanks, i will | 07:07 |
| Haomeng | jiangfei: thanks | 07:07 |
| Haomeng | jiangfei: if our ironic driver-list can not list all dist-conductor's drivers, that should be another bug I think:) | 07:11 |
| jiangfei | Haomeng:i will test it | 07:13 |
| Haomeng | jiangfei: we can check the code logic, that we will know if it support multi-conductors | 07:14 |
| Haomeng | jiangfei: and we can levreage the logic to do pre-check for node-craete -d xxx | 07:14 |
| *** k4n0 has joined #openstack-ironic | 07:15 | |
| jiangfei | Haomeng:ok | 07:17 |
| Haomeng | tjiangfei: hank you | 07:17 |
| jiangfei | Haomeng: :) | 07:18 |
| Haomeng | jiangfei: :) | 07:18 |
| *** Masahiro has joined #openstack-ironic | 07:33 | |
| *** Masahiro has quit IRC | 07:37 | |
| *** ifarkas has joined #openstack-ironic | 07:40 | |
| *** mrda is now known as mrda-away | 08:07 | |
| *** Nisha has quit IRC | 08:16 | |
| *** jistr has joined #openstack-ironic | 08:20 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-specs: Set blueprint link for drivers capabilities spec https://review.openstack.org/140267 | 08:26 |
| *** dtantsur|afk is now known as dtantsur | 08:27 | |
| dtantsur | Morning | 08:27 |
| *** ndipanov_gone is now known as ndipanov | 08:29 | |
| dtantsur | devananda, https://review.openstack.org/140267 <-- blueprint for driver capabilities | 08:31 |
| *** jcoufal has joined #openstack-ironic | 08:33 | |
| *** Masahiro has joined #openstack-ironic | 08:33 | |
| *** MattMan has quit IRC | 08:33 | |
| dtantsur | ifarkas, o/ once you have some time, could you start reviewing a long (6 for now) patch queue starting with https://review.openstack.org/#/c/140064/ ? | 08:35 |
| *** MattMan has joined #openstack-ironic | 08:36 | |
| *** Masahiro has quit IRC | 08:38 | |
| ifarkas | dtantsur, sure, will do ;-) | 08:39 |
| dtantsur | cool! | 08:39 |
| * dtantsur did some refactoring yesterday | 08:39 | |
| dtantsur | ifarkas, fun thing: zuul was so surprised with my 4 patches uploaded yesterday, that I have to use 'recheck' today to make it run tests on them :) | 08:41 |
| ifarkas | dtantsur, that must have been a nice surprise ;-) | 08:42 |
| dtantsur | heh | 08:42 |
| rameshg87 | ifarkas, hi | 08:43 |
| ifarkas | rameshg87, hi | 08:43 |
| rameshg87 | ifarkas, there was lot of discussions yesterday on RAID configuration. had a chance to see them ? | 08:43 |
| rameshg87 | ifarkas, i think we should post a full spec as all agreed this is in scope | 08:44 |
| ifarkas | rameshg87, I haven't yet, but I will read back the log | 08:45 |
| rameshg87 | ifarkas, okay. i am think of addressing the comments in the spec, but wanted to discuss with you | 08:47 |
| rameshg87 | ifarkas, let me know once you have followed up on what has happened, then we can discuss | 08:47 |
| ifarkas | rameshg87, ok | 08:48 |
| GheRivero | morning all | 08:48 |
| rameshg87 | morning GheRivero | 08:49 |
| dtantsur | GheRivero, rameshg87, o/ | 08:52 |
| rameshg87 | dtantsur, morning o/ | 08:52 |
| *** Haomeng|2 has joined #openstack-ironic | 08:52 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Enable iscsi_ilo driver to share boot images https://review.openstack.org/137291 | 08:53 |
| rameshg87 | dtantsur, when/how is a short-spec accepted for current release ? | 08:53 |
| rameshg87 | dtantsur, by getting some +1s from cores ? | 08:53 |
| *** Haomeng has quit IRC | 08:54 | |
| *** jcoufal_ has joined #openstack-ironic | 08:55 | |
| dtantsur | rameshg87, short specs are accepted to backlog or converted directly to long specs after seeing some consensus on the idea | 08:57 |
| dtantsur | rameshg87, in your case it makes sense to convert it to long spec and retarget to specs/kilo | 08:57 |
| dtantsur | brb | 08:57 |
| rameshg87 | dtantsur, okay | 08:58 |
| *** jcoufal has quit IRC | 08:59 | |
| *** dlpartain has joined #openstack-ironic | 09:08 | |
| *** dividebin has quit IRC | 09:11 | |
| *** dividehex has joined #openstack-ironic | 09:14 | |
| *** romcheg has joined #openstack-ironic | 09:19 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Add support for VirtualBox WebService. https://review.openstack.org/137926 | 09:24 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Add support for VirtualBox WebService. https://review.openstack.org/137926 | 09:26 |
| *** athomas has joined #openstack-ironic | 09:28 | |
| *** derekh has joined #openstack-ironic | 09:35 | |
| *** erwan_taf has joined #openstack-ironic | 09:37 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Enable boot images sharing for iLO driver(s) https://review.openstack.org/137291 | 09:41 |
| *** andreykurilin_ has joined #openstack-ironic | 09:42 | |
| rameshg87 | dtantsur, can you please have a look at some of my reviews when you get time, some them have been waiting without feedback, posted the full specs - https://review.openstack.org/137291, https://review.openstack.org/137567, https://review.openstack.org/137926 | 09:42 |
| *** rameshg87 is now known as rameshg87-away | 09:43 | |
| *** lazy_prince is now known as killer_prince | 09:46 | |
| *** igordcard has joined #openstack-ironic | 09:47 | |
| *** pelix has joined #openstack-ironic | 09:48 | |
| dtantsur | rameshg87-away, yeah, will try to find time today | 09:48 |
| *** andreykurilin_ has quit IRC | 09:49 | |
| *** andreykurilin_ has joined #openstack-ironic | 09:49 | |
| *** killer_prince is now known as lazy_prince | 09:49 | |
| *** dtantsur is now known as dtantsur|brb | 09:49 | |
| *** dlpartain has quit IRC | 09:50 | |
| *** andreykurilin__ has joined #openstack-ironic | 10:00 | |
| *** andreykurilin_ has quit IRC | 10:00 | |
| *** lazy_prince is now known as killer_prince | 10:07 | |
| *** chenglch has quit IRC | 10:18 | |
| *** vdrok has joined #openstack-ironic | 10:19 | |
| *** jistr has quit IRC | 10:21 | |
| *** Masahiro has joined #openstack-ironic | 10:22 | |
| *** Marga_ has joined #openstack-ironic | 10:24 | |
| *** Masahiro has quit IRC | 10:27 | |
| openstackgerrit | Merged stackforge/ironic-discoverd: Wait for power off state before calling discovery done https://review.openstack.org/140064 | 10:30 |
| *** erwan_taf has quit IRC | 10:30 | |
| *** erwan_taf has joined #openstack-ironic | 10:33 | |
| *** marcoemorais1 has quit IRC | 10:40 | |
| *** killer_prince is now known as lazy_prince | 10:41 | |
| *** wanghua has joined #openstack-ironic | 10:41 | |
| *** ramineni has quit IRC | 10:43 | |
| openstackgerrit | Merged stackforge/ironic-discoverd: Refactoring: consolidate standard plugins in one module https://review.openstack.org/140134 | 10:45 |
| openstackgerrit | Merged stackforge/ironic-discoverd: Refactoring: drop features incompatible with Kilo changes https://review.openstack.org/140135 | 10:45 |
| *** andreykurilin__ has quit IRC | 10:48 | |
| *** aignatov has quit IRC | 10:48 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Do not strip 'glance://' prefix from image hrefs https://review.openstack.org/139057 | 10:48 |
| *** jistr has joined #openstack-ironic | 10:49 | |
| *** aignatov has joined #openstack-ironic | 10:49 | |
| sambetts | morning dtantsur | 10:59 |
| rameshg87-away | dtantsur|brb, thanks | 11:07 |
| * rameshg87-away is going home | 11:07 | |
| *** rameshg87-away has quit IRC | 11:08 | |
| *** alexpilotti has joined #openstack-ironic | 11:17 | |
| *** Marga_ has quit IRC | 11:17 | |
| *** enterprisedc has joined #openstack-ironic | 11:23 | |
| *** pradipta_away is now known as pradipta | 11:32 | |
| pensu | Hey all, can devstack use the Ironic's external DHCP feature? Ref: https://blueprints.launchpad.net/devstack/+spec/ironic-dnsmasq-external-dhcp | 11:36 |
| *** pradipta is now known as pradipta_away | 11:38 | |
| *** dtantsur|brb is now known as dtantsur | 11:46 | |
| dtantsur | sambetts, o/ | 11:46 |
| *** masco has joined #openstack-ironic | 11:56 | |
| *** pensu has quit IRC | 12:05 | |
| *** Masahiro has joined #openstack-ironic | 12:12 | |
| *** Masahiro has quit IRC | 12:15 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Do not strip 'glance://' prefix from image hrefs https://review.openstack.org/139057 | 12:17 |
| *** Marga_ has joined #openstack-ironic | 12:18 | |
| *** bradjones has joined #openstack-ironic | 12:20 | |
| *** Marga_ has quit IRC | 12:23 | |
| *** smoriya has quit IRC | 12:23 | |
| *** wanghua has quit IRC | 12:44 | |
| openstackgerrit | Arata Notsu proposed openstack/ironic: Correct vmware ssh power manager https://review.openstack.org/139810 | 12:45 |
| *** alexpilotti has quit IRC | 12:46 | |
| *** rameshg87 has joined #openstack-ironic | 12:55 | |
| rameshg87 | vdrok, hi | 13:01 |
| rameshg87 | vdrok, is there some bigger use-case that you have in mind rather than allowing to *also* mention glance://uuid instead of just uuid | 13:01 |
| rameshg87 | ifarkas, hi | 13:01 |
| *** rakesh_hs has quit IRC | 13:02 | |
| ifarkas | rameshg87, hi | 13:03 |
| rameshg87 | ifarkas, got a chance to read the backlog ? :) | 13:03 |
| ifarkas | rameshg87, yeah, I did that | 13:03 |
| rameshg87 | ifarkas, okay to discuss comments right now ? | 13:04 |
| ifarkas | rameshg87, can we do it later? I am in the middle of something | 13:05 |
| rameshg87 | ifarkas, sure .. | 13:05 |
| ifarkas | rameshg87, thanks | 13:05 |
| *** dprince has joined #openstack-ironic | 13:08 | |
| *** Marga_ has joined #openstack-ironic | 13:19 | |
| *** foexle has joined #openstack-ironic | 13:19 | |
| *** Marga_ has quit IRC | 13:23 | |
| *** alexpilotti has joined #openstack-ironic | 13:25 | |
| vdrok | rameshg87, hi :) | 13:39 |
| vdrok | rameshg87, it's about this spec - https://review.openstack.org/#/c/135276/ | 13:39 |
| rameshg87 | vdrok, i was actually wondering on rloo's statement | 13:39 |
| rameshg87 | vdrok, ah okay | 13:40 |
| vdrok | rameshg87, at some point they should be just some url | 13:40 |
| rameshg87 | vdrok, then why not mention the blueprint as part of the submittal | 13:40 |
| rameshg87 | vdrok, i mean as part of commit msg | 13:40 |
| vdrok | rameshg87, idk, i can add this, but it's mentioned in dependency | 13:41 |
| vdrok | rameshg87, you mean implements bp ? | 13:41 |
| rameshg87 | vdrok, yes | 13:42 |
| rameshg87 | vdrok, i think that makes things a lot clear | 13:42 |
| rameshg87 | vdrok, as to why we are doing this | 13:42 |
| vdrok | rameshg87, ok, will update | 13:42 |
| rameshg87 | vdrok, thanks | 13:44 |
| vdrok | rameshg87, np :) | 13:44 |
| rameshg87 | vdrok, and i did see you mentioned iLO in that review link https://review.openstack.org/#/c/139057/ | 13:44 |
| *** rushiagr is now known as rushiagr_away | 13:44 | |
| vdrok | rameshg87, yup, its just cosmetic change for now | 13:45 |
| vdrok | rameshg87, just to not include it in the next one | 13:45 |
| rameshg87 | vdrok, okay | 13:46 |
| *** linggao has joined #openstack-ironic | 13:46 | |
| rameshg87 | vdrok, let me know if you need some info in that space :) | 13:46 |
| vdrok | rameshg87, ok, thanks :) actually there is one thing | 13:47 |
| vdrok | rameshg87, can ilovirtualmediaiscsi be used with gen7 server? | 13:47 |
| vdrok | rameshg87, or gen8 and ilo4 are minimal requirements? | 13:47 |
| rameshg87 | vdrok, there are no features that we are using that are *only* available in gen8 | 13:48 |
| rameshg87 | vdrok, but we didn't test it on gen7 | 13:48 |
| vdrok | rameshg87, ok, will try then at some point :) thanks | 13:49 |
| rameshg87 | vdrok, somethings in the dependent library proliantutils are broken https://github.com/hpproliant/proliantutils/ | 13:49 |
| rameshg87 | vdrok, because gen7 ilo3 doesn't understand everything in that module | 13:49 |
| rameshg87 | vdrok, we are in the process of fixing them actually | 13:49 |
| vdrok | rameshg87, i see, great | 13:49 |
| rameshg87 | vdrok, will update you once we fix them if you are interested | 13:49 |
| vdrok | rameshg87, thanks | 13:50 |
| openstackgerrit | Syed Ismail Faizan Barmawer proposed openstack/ironic-specs: Partition image support for agent driver https://review.openstack.org/137363 | 13:51 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Do not strip 'glance://' prefix from image hrefs https://review.openstack.org/139057 | 13:52 |
| *** Masahiro has joined #openstack-ironic | 14:00 | |
| *** Masahiro has quit IRC | 14:04 | |
| *** rloo has joined #openstack-ironic | 14:04 | |
| *** Marga_ has joined #openstack-ironic | 14:06 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Refactoring: split test.py https://review.openstack.org/140339 | 14:16 |
| *** k4n0 has quit IRC | 14:21 | |
| *** Marga_ has quit IRC | 14:22 | |
| *** Marga_ has joined #openstack-ironic | 14:22 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Refactoring: split test.py https://review.openstack.org/140339 | 14:26 |
| jroll | morning, all | 14:26 |
| *** ryanpetrello has joined #openstack-ironic | 14:26 | |
| dtantsur | jroll, morning | 14:26 |
| rloo | morning jroll, afternoon dtantsur | 14:27 |
| dtantsur | rloo, hi :) | 14:27 |
| *** Marga_ has quit IRC | 14:27 | |
| jroll | hey rloo and dtantsur :) | 14:28 |
| vdrok | morning jroll, rloo and dtantsur | 14:31 |
| jroll | \o vdrok | 14:31 |
| rloo | hey, afternoon vdrok | 14:32 |
| dtantsur | vdrok, o/ | 14:32 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Refactoring: split test.py https://review.openstack.org/140339 | 14:34 |
| victor_lowther | goood morning | 14:36 |
| dtantsur | victor_lowther, hi! | 14:36 |
| * victor_lowther is caffinating | 14:36 | |
| jroll | JoshNang: this needs a rebase: https://review.openstack.org/#/c/92627/ | 14:38 |
| *** athomas has quit IRC | 14:39 | |
| NobodyCam | good morning Ironic | 14:40 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Remove python-tox from dev-quickstart docs https://review.openstack.org/139223 | 14:42 |
| jroll | morning NobodyCam | 14:42 |
| NobodyCam | morning jroll | 14:42 |
| *** athomas has joined #openstack-ironic | 14:47 | |
| *** rushiagr_away is now known as rushiagr | 14:48 | |
| *** alexpilotti_ has joined #openstack-ironic | 14:54 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add AMT-PXE-Driver to deploy cloud on PC https://review.openstack.org/135184 | 14:57 |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add AMT Support with PXE Driver https://review.openstack.org/140351 | 14:57 |
| *** alexpilotti has quit IRC | 14:57 | |
| *** alexpilotti_ is now known as alexpilotti | 14:57 | |
| *** ndipanov has quit IRC | 14:59 | |
| *** ndipanov has joined #openstack-ironic | 14:59 | |
| * dtantsur sees Fuel Agent as a nice holy war topic for the next meeting :) | 14:59 | |
| *** rushiagr is now known as rushiagr_away | 15:00 | |
| NobodyCam | dtantsur: tonights meeting? | 15:00 |
| * jroll just sent another email about that | 15:01 | |
| *** zz_jgrimm is now known as jgrimm | 15:01 | |
| jroll | NobodyCam: it was last night :P | 15:01 |
| dtantsur | NobodyCam, oh, meeting is tonight? I thought it _was_ tonight, no? | 15:01 |
| jroll | it was tuesday 5am UTC | 15:01 |
| NobodyCam | 0500 UTC on Tuesday | 15:01 |
| jroll | right now is tuesday 1500 UTC | 15:01 |
| *** k4n0 has joined #openstack-ironic | 15:01 | |
| NobodyCam | d'oh | 15:01 |
| jroll | :) | 15:01 |
| dtantsur | aha good :) | 15:01 |
| jroll | check out the meeting minutes, we made a plan for the state machine spec | 15:02 |
| jroll | or the logs, I guess | 15:02 |
| dtantsur | awesome \o/ | 15:02 |
| *** pawel-palucki has joined #openstack-ironic | 15:02 | |
| *** shakamunyi has joined #openstack-ironic | 15:04 | |
| *** shakamunyi has quit IRC | 15:04 | |
| *** shakamunyi has joined #openstack-ironic | 15:05 | |
| *** jgrimm is now known as zz_jgrimm | 15:08 | |
| *** lazy_prince is now known as killer_prince | 15:11 | |
| NobodyCam | wow awesome plan.. | 15:12 |
| *** rushiagr_away is now known as rushiagr | 15:14 | |
| *** ChuckC_ has joined #openstack-ironic | 15:15 | |
| *** ChuckC has quit IRC | 15:15 | |
| *** Haomeng has joined #openstack-ironic | 15:20 | |
| *** ChuckC_ has quit IRC | 15:21 | |
| *** Haomeng|2 has quit IRC | 15:22 | |
| *** Marga_ has joined #openstack-ironic | 15:23 | |
| *** shakamunyi has quit IRC | 15:25 | |
| *** shakamunyi has joined #openstack-ironic | 15:25 | |
| rloo | jroll: I just read the meeting minutes. I didn't see anything about state machine logs. Oh wait, you meant look at the meeting logs ;) | 15:27 |
| *** Marga_ has quit IRC | 15:28 | |
| jroll | >.> | 15:28 |
| NobodyCam | rloo: it starts around half past the hour | 15:28 |
| NobodyCam | and morning | 15:28 |
| rloo | NobodyCam: yeah. I read that. I thought they had a plan for 'logs' for the state machine. I didn't know what that meant ;) | 15:29 |
| NobodyCam | :) | 15:29 |
| rloo | and morning to you NobodyCam and victor_lowther ;) | 15:29 |
| *** dlpartain has joined #openstack-ironic | 15:30 | |
| *** alexm__ has joined #openstack-ironic | 15:31 | |
| rloo | btw, victor_lowther, wrt the state machine spec, it'd be nice to put something at the top about how it is a spec describing the intention, not the implementation. | 15:31 |
| *** zz_jgrimm is now known as jgrimm | 15:32 | |
| victor_lowther | Yes. | 15:32 |
| victor_lowther | I am updating it now. | 15:32 |
| rloo | thx victor_lowther! | 15:32 |
| victor_lowther | Spam me with your asks, lest things get +2'ed without your feedback. :) | 15:32 |
| *** ndipanov_ has joined #openstack-ironic | 15:33 | |
| *** jerryz has quit IRC | 15:33 | |
| *** ndipanov has quit IRC | 15:34 | |
| alexm__ | hello guys, does Ironic support multiple disks per node? Conductor will alwyas chose the first disk I presume? | 15:34 |
| jroll | alexm__: correct | 15:35 |
| rloo | victor_lowther: I think the wolves are hungry and ready to pounce ... ;) We can always update/refine the wording if need be, later. I just want to make sure someone new reading it, understands the context. | 15:35 |
| *** trown has quit IRC | 15:35 | |
| jroll | alexm__: there's work being done to make it pick other disks, see https://review.openstack.org/#/c/138729/ | 15:35 |
| *** trown has joined #openstack-ironic | 15:36 | |
| alexm__ | jroll: thanks. HW Raid 1 should work seemlessly too? | 15:37 |
| jroll | alexm__: if you set it up ahead of time, I think it should... ironic does not configure RAID today | 15:38 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Remove duplicate dependencies from dev-quickstart docs https://review.openstack.org/139223 | 15:45 |
| jroll | ^ easy review | 15:45 |
| *** Masahiro has joined #openstack-ironic | 15:48 | |
| *** ryanpetrello has quit IRC | 15:49 | |
| Shrews | jroll: one suggestion there... something i've been meaning to add myself | 15:49 |
| jroll | nice, good call | 15:50 |
| * dtantsur brb | 15:50 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Remove duplicate dependencies from dev-quickstart docs https://review.openstack.org/139223 | 15:51 |
| jroll | Shrews: fixed | 15:51 |
| jroll | thanks for that | 15:51 |
| Shrews | jroll: thx. i'm not 100% certain that nose is even needed anymore, but lgtm now | 15:52 |
| jroll | agree, but didn't want to look :P | 15:52 |
| *** Masahiro has quit IRC | 15:53 | |
| *** ChuckC_ has joined #openstack-ironic | 15:54 | |
| NobodyCam | brb | 15:56 |
| *** ChuckC_ has quit IRC | 15:59 | |
| *** ndipanov_ has quit IRC | 16:01 | |
| *** jcoufal_ has quit IRC | 16:02 | |
| *** dlaube has joined #openstack-ironic | 16:03 | |
| *** masco has quit IRC | 16:03 | |
| *** pcrews has joined #openstack-ironic | 16:06 | |
| *** ChuckC_ has joined #openstack-ironic | 16:12 | |
| *** ChuckC_ is now known as ChuckC | 16:13 | |
| *** ndipanov_ has joined #openstack-ironic | 16:13 | |
| Shrews | I wonder if the state machine spec should call out changes needed for the nova driver.... | 16:14 |
| NobodyCam | JayF: quick question. were you going to add that log line to 92627? | 16:17 |
| rloo | Shrews: I thought it mentioned something about the nova driver? | 16:17 |
| *** achanda has joined #openstack-ironic | 16:17 | |
| Shrews | rloo: oh yes... so short i missed it | 16:18 |
| rloo | Shrews, jroll: either of you test 139223 for opensuse/sle? | 16:23 |
| *** Marga_ has joined #openstack-ironic | 16:23 | |
| Shrews | rloo: not I | 16:24 |
| openstackgerrit | Victor Lowther proposed openstack/ironic-specs: New Ironic provisioner state machine. https://review.openstack.org/133828 | 16:26 |
| NobodyCam | woo hoo | 16:26 |
| jroll | rloo: no... don't see any reason why it wouldn't work | 16:26 |
| rloo | jroll: famous last words ;) I'll approve anyway. | 16:26 |
| jroll | ha, ok | 16:26 |
| jroll | thanks | 16:26 |
| *** erwan_taf has quit IRC | 16:26 | |
| JayF | NobodyCam: I personally was not, because at that point all three "j*" would have code in it. If you put the comment on it I'm sure JoshNang will | 16:28 |
| *** Marga_ has quit IRC | 16:28 | |
| NobodyCam | JayF: ack | 16:28 |
| NobodyCam | will do | 16:28 |
| *** enikanorov has quit IRC | 16:30 | |
| *** enikanorov has joined #openstack-ironic | 16:30 | |
| *** ryanpetrello has joined #openstack-ironic | 16:30 | |
| NobodyCam | JayF: done | 16:31 |
| *** subscope has quit IRC | 16:32 | |
| JayF | adam_g: https://review.openstack.org/139770 is landing \o/ (your devstack fix) | 16:32 |
| JoshNang | NobodyCam: ahh good call. Will add today | 16:32 |
| * jroll steps away for a while | 16:32 | |
| NobodyCam | :) | 16:32 |
| *** shakamunyi has quit IRC | 16:33 | |
| *** k4n0 has quit IRC | 16:34 | |
| devananda | g'morning, all | 16:35 |
| dtantsur | devananda, morning | 16:35 |
| NobodyCam | good morning devananda | 16:36 |
| NobodyCam | oh btw state machine has 2 +2's | 16:36 |
| dtantsur | land LAND :D | 16:37 |
| victor_lowther | well, there is a spelling misteak to fix. | 16:38 |
| *** dlpartain has quit IRC | 16:38 | |
| NobodyCam | :-p | 16:38 |
| Shrews | misteaks are nawt aloud | 16:39 |
| NobodyCam | lol | 16:40 |
| * devananda reads the fuel thread before coffee while wondering if that's a wise thing to do | 16:40 | |
| NobodyCam | devananda: I am on second cup and still not through it... | 16:41 |
| NobodyCam | lol it keeps growing as I read it | 16:41 |
| *** ryanpetrello has quit IRC | 16:42 | |
| * dtantsur refrains for some time to get the other folks' opinions | 16:42 | |
| * Shrews has just realized that his misconfigured mail filter has deleted that entire thread | 16:44 | |
| victor_lowther | and Jenkins hates my table. :( | 16:44 |
| *** dlpartain has joined #openstack-ironic | 16:44 | |
| NobodyCam | brb | 16:44 |
| dtantsur | Shrews, good filter :D | 16:45 |
| Shrews | lol | 16:45 |
| *** krtaylor has quit IRC | 16:46 | |
| *** shakamunyi has joined #openstack-ironic | 16:47 | |
| rloo | victor_lowther: I don't know if that table is necessary. I was asking about the verbs, to try to pin down what we actually will end up calling them. but that is out of scope from this spec anyway. | 16:47 |
| *** dlpartain has quit IRC | 16:47 | |
| rloo | victor_lowther: although it is a nice table ;) | 16:48 |
| *** Marga_ has joined #openstack-ironic | 16:59 | |
| *** achanda has quit IRC | 17:01 | |
| *** Marga_ has quit IRC | 17:03 | |
| NobodyCam | I like the table :) | 17:07 |
| *** achanda has joined #openstack-ironic | 17:08 | |
| *** krtaylor has joined #openstack-ironic | 17:09 | |
| *** ifarkas is now known as ifarkas_afk | 17:10 | |
| devananda | wow - 19 comments in under an hour? | 17:10 |
| devananda | I think that's a record for one of our specs :) | 17:10 |
| devananda | rloo: as far as marking the spec informational, I will also do that in launchpad | 17:11 |
| devananda | rloo: when we tag it as completed, I will close it with the status "informational" rather than "implemented" | 17:11 |
| devananda | ooh. table of verbs. nice | 17:12 |
| *** datajerk has joined #openstack-ironic | 17:12 | |
| *** ryanpetrello has joined #openstack-ironic | 17:13 | |
| *** masco has joined #openstack-ironic | 17:13 | |
| devananda | rloo, victor_lowther: fwiw, I think the table provides a helpful reference. certainly for me, as I get to implementing that in the state machine | 17:15 |
| *** afaranha has joined #openstack-ironic | 17:15 | |
| devananda | it's easier to read and compare my code to, so I'm ++ on leaving it in the spec | 17:15 |
| NobodyCam | ++ for leaving the table in the spec | 17:17 |
| *** linggao has quit IRC | 17:17 | |
| *** linggao has joined #openstack-ironic | 17:18 | |
| rloo | devananda: ok, if the table is useful for you ;) | 17:18 |
| openstackgerrit | ZhiQiang Fan proposed openstack/ironic: rename oslo.concurrency to oslo_concurrency https://review.openstack.org/139782 | 17:18 |
| JayF | is ^ that person on irc? | 17:19 |
| rloo | devananda: the table seemed like a diff representation (but same info) than what was in the diagram | 17:19 |
| *** rameshg87 has quit IRC | 17:19 | |
| devananda | rloo: yup | 17:19 |
| openstackgerrit | Victor Lowther proposed openstack/ironic-specs: New Ironic provisioner state machine. https://review.openstack.org/133828 | 17:19 |
| rloo | devananda: maybe more useful to have the table under the diagram? | 17:20 |
| JayF | https://review.openstack.org/#/c/138116/ this is passing all tests and awaiting review | 17:20 |
| victor_lowther | That should have a table that Jenkins will not whine about. | 17:21 |
| JoshNang | JayF: will review | 17:23 |
| JayF | JoshNang: doesn't it make you warm and fuzzy to see the tempest job voting there | 17:24 |
| JayF | lol | 17:24 |
| JoshNang | JayF: ha very much so | 17:24 |
| *** marcoemorais has joined #openstack-ironic | 17:26 | |
| JayF | Where are Ironic meeting logs kept? | 17:26 |
| *** pensu has joined #openstack-ironic | 17:27 | |
| JayF | sadly doesn't appear linked from https://wiki.openstack.org/wiki/Meetings/Ironic | 17:27 |
| devananda | JayF: should be there ... | 17:28 |
| NobodyCam | JayF: http://eavesdrop.openstack.org/meetings/ironic/2014 | 17:28 |
| devananda | JayF: "Previos Meetings" | 17:28 |
| NobodyCam | link is on the bottom of the agenda | 17:28 |
| * devananda updates the agenda after last night's meeting | 17:29 | |
| NobodyCam | lol /me was all ready for the meeting tonight :( | 17:29 |
| JayF | d'oh | 17:30 |
| devananda | NobodyCam: :( | 17:30 |
| openstackgerrit | Merged openstack/ironic: Remove duplicate dependencies from dev-quickstart docs https://review.openstack.org/139223 | 17:31 |
| NobodyCam | hehe it was my bad.. | 17:31 |
| JoshNang | so, with the split between cleaning and zapping, should they be separate specs? i'd lean towards yes, with zapping being dependent on cleaning | 17:32 |
| *** penick has joined #openstack-ironic | 17:32 | |
| devananda | victor_lowther: tabs? | 17:33 |
| JayF | JoshNang: I agree | 17:33 |
| victor_lowther | That is how I would do it | 17:33 |
| devananda | victor_lowther: 12 to 13 introduced a lot of tabs :( | 17:34 |
| devananda | https://review.openstack.org/#/c/133828/12..13/specs/kilo/new-ironic-state-machine.rst | 17:34 |
| *** ChuckC has quit IRC | 17:34 | |
| victor_lowther | (and deploying as well for that matter) | 17:34 |
| victor_lowther | devananda: I blame Emacs on a machine I have not finished setting up. | 17:34 |
| devananda | heh | 17:35 |
| devananda | victor_lowther: you had me at "i blame emacs" | 17:35 |
| victor_lowther | so it did the Wrong Thing with M-x whitespace-cleanup | 17:35 |
| NobodyCam | the tabs actually looked ok in the rendered version: http://docs-draft.openstack.org/28/133828/13/check/gate-ironic-specs-docs/01f36dd/doc/build/html/specs/kilo/new-ironic-state-machine.html | 17:35 |
| rloo | maybe we need a unit test to check for tabs :-) | 17:37 |
| *** Masahiro has joined #openstack-ironic | 17:37 | |
| NobodyCam | are tabs a bad for non code docs? | 17:38 |
| victor_lowther | detabbing is easy enough | 17:41 |
| NobodyCam | should we set the spec url on https://blueprints.launchpad.net/ironic/+spec/new-ironic-state-machine | 17:42 |
| *** Masahiro has quit IRC | 17:42 | |
| *** spandhe has joined #openstack-ironic | 17:43 | |
| victor_lowther | NobodyCam: done. | 17:43 |
| NobodyCam | :) | 17:43 |
| devananda | NobodyCam: tabs in general are bad for formatting of text-based documents, whether code or otherwise | 17:43 |
| NobodyCam | ack :) | 17:44 |
| *** Marga_ has joined #openstack-ironic | 17:45 | |
| *** ChuckC has joined #openstack-ironic | 17:47 | |
| *** ryanpetrello has quit IRC | 17:52 | |
| *** ryanpetrello has joined #openstack-ironic | 17:54 | |
| openstackgerrit | Victor Lowther proposed openstack/ironic-specs: New Ironic provisioner state machine. https://review.openstack.org/133828 | 17:55 |
| victor_lowther | Now with no tabs and rloo's nits picked. | 17:55 |
| victor_lowther | and now for lunch. | 17:56 |
| *** igordcard has quit IRC | 17:57 | |
| NobodyCam | need to run out and grab some mile for coffee... bbiafm | 17:58 |
| *** david-lyle_afk is now known as david-lyle | 17:58 | |
| *** derekh has quit IRC | 18:00 | |
| *** dtantsur is now known as dtantsur|afk | 18:05 | |
| *** wanyen has joined #openstack-ironic | 18:05 | |
| wanyen | Hi Deva | 18:06 |
| *** masco has quit IRC | 18:07 | |
| wanyen | Hi devananda, are you there? I would like to talk to you about https://review.openstack.org/#/c/136104/ | 18:07 |
| devananda | wanyen: hi! it has a +2 from nova PTL - what would you like to talk with me about? | 18:09 |
| wanyen | we need this spec to support a node with multiple caps, e.g., secure boot, uefi, bios and tboot. | 18:09 |
| devananda | sure | 18:09 |
| wanyen | can you talk to Nova PTL when you attend next Nova meeting | 18:10 |
| devananda | wanyen: about what? it already has a +2 from him | 18:10 |
| wanyen | I would like to we need anothe +2 | 18:10 |
| devananda | wanyen: it's not his responsibility to get other +2's | 18:10 |
| *** rushiagr is now known as rushiagr_away | 18:12 | |
| devananda | wanyen: you should ask in #openstack-nova or attend their meetings. You could also post the code for review now, so that nova reviewers can see that it is a small change | 18:12 |
| wanyen | Are there any ironic core reviewere also Nova core reviewer? | 18:12 |
| devananda | wanyen: not afaik, no | 18:12 |
| wanyen | Deva: okay. We will. | 18:13 |
| devananda | it's hard to maintain that level of activity in two active projects ... | 18:13 |
| *** athomas has quit IRC | 18:13 | |
| wanyen | The change is needed in Nova ironic driver, so ironic own the implementation of that driver. Right? | 18:14 |
| devananda | wanyen: while we are responsible for maintaining it, nova-core is responsible for reviewing and approving changes to it | 18:14 |
| wanyen | deva, yes I understand. | 18:15 |
| wanyen | Are there any changes required in Nova Ironic driver, e.g., state machine changes? | 18:16 |
| wanyen | s/any changes/any other changes | 18:17 |
| devananda | possibly. I'm going to try to minimize any impact on Nova through changes in our REST API | 18:17 |
| wanyen | deva, I thinking we get all the Nova ironic driver changes in one code review assuming specs can be approved soon. What do you think? | 18:19 |
| devananda | wanyen: I actually dont think that's a good idea. Nova-core has made it clear that they prefer small, well contained, changes -- these are easier for them to review | 18:20 |
| devananda | wanyen: if I understand the spec you referenced above, I think the change there will be very small | 18:20 |
| devananda | not counting tests, I would expect it to be less than 100 lines | 18:20 |
| *** romcheg has quit IRC | 18:20 | |
| * NobodyCam is back | 18:20 | |
| *** romcheg has joined #openstack-ironic | 18:21 | |
| wanyen | deva, okay. In that case, we will submit code changes for https://review.openstack.org/#/c/136104/ as soon as we can. The changes should be small. | 18:21 |
| devananda | if that's correct, and it doesn't affect anything outside of Ironic, it will probably land easily | 18:21 |
| devananda | nova core give a lot of leeway to changes within drivers -- as long as they don't impact any other part of nova | 18:21 |
| devananda | so an approved spec with priority:None is OK | 18:21 |
| devananda | it means "we'll get to it, if we have time" -- and a small change takes less time to review :) | 18:21 |
| wanyen | deva, good. | 18:22 |
| wanyen | deva, tx | 18:22 |
| devananda | wanyen: np | 18:23 |
| *** ndipanov_ is now known as ndipanov_gone | 18:23 | |
| *** harlowja_away is now known as harlowja | 18:30 | |
| *** achanda has quit IRC | 18:34 | |
| *** Marga_ has quit IRC | 18:35 | |
| *** r-daneel has joined #openstack-ironic | 18:35 | |
| *** Marga_ has joined #openstack-ironic | 18:36 | |
| *** shakamunyi has quit IRC | 18:39 | |
| *** Marga_ has quit IRC | 18:41 | |
| *** jistr has quit IRC | 18:49 | |
| *** pelix has quit IRC | 18:53 | |
| *** spandhe has quit IRC | 18:57 | |
| *** ifarkas_afk has quit IRC | 18:57 | |
| *** rushiagr_away is now known as rushiagr | 18:59 | |
| *** andreykurilin_ has joined #openstack-ironic | 18:59 | |
| *** Nisha has joined #openstack-ironic | 19:01 | |
| *** spandhe has joined #openstack-ironic | 19:04 | |
| *** Nisha_away has joined #openstack-ironic | 19:07 | |
| *** Nisha has quit IRC | 19:10 | |
| *** Marga_ has joined #openstack-ironic | 19:11 | |
| *** Marga_ has quit IRC | 19:14 | |
| *** Nisha_away has quit IRC | 19:15 | |
| *** Nisha has joined #openstack-ironic | 19:15 | |
| *** pensu has quit IRC | 19:16 | |
| *** bradjones has quit IRC | 19:16 | |
| NobodyCam | second +2 added to new state machine spec. not +A'd just so others can re-review | 19:20 |
| *** Marga_ has joined #openstack-ironic | 19:21 | |
| * NobodyCam wounders if it would be worth adding an "affected states" section to the spec template (for next cycle, not this one) | 19:24 | |
| Nisha | NobodyCam, devananda , jroll, JayF ....just to make sure, since the state for introspection is now INSPECTING and INSPECTFAIL, do we need to have the end point as inspect or introspect? | 19:25 |
| Nisha | #link https://review.openstack.org/#/c/100951/50 | 19:25 |
| NobodyCam | Nisha: line 304 of https://review.openstack.org/#/c/133828/14/specs/kilo/new-ironic-state-machine.rst | 19:26 |
| jroll | inspect sounds right | 19:26 |
| NobodyCam | inspect | 19:26 |
| *** Masahiro has joined #openstack-ironic | 19:26 | |
| Nisha | jroll, NobodyCam thanks | 19:26 |
| NobodyCam | :) | 19:26 |
| Shrews | i really hope someone makes an inspector gadget class | 19:26 |
| NobodyCam | lol | 19:27 |
| NobodyCam | gogo Shrews | 19:27 |
| Nisha | the CLI also to be named as node-inspect? | 19:27 |
| jroll | +1 | 19:28 |
| Nisha | thanks jroll | 19:28 |
| Nisha | NobodyCam, jroll i had one more ques | 19:28 |
| NobodyCam | sure | 19:28 |
| *** bradjones has joined #openstack-ironic | 19:29 | |
| *** Marga_ has quit IRC | 19:29 | |
| Nisha | I have a working code for introspection ready (some minor changes to be done as per state spec now), but the code depends on the new states. For now i have introduced the states in my code changes | 19:29 |
| Nisha | So for posting the code for review do i need to wait for state code to be available first? | 19:30 |
| Nisha | or for now i can submit the code changes as i have them today | 19:30 |
| NobodyCam | Nisha: have you seen devananda's FSM work. https://review.openstack.org/#/c/139215/ | 19:30 |
| jroll | you can put the code up :) | 19:30 |
| *** Masahiro has quit IRC | 19:30 | |
| Nisha | and then later replace the states for inspection from states spec | 19:31 |
| *** rushiagr is now known as rushiagr_away | 19:31 | |
| Nisha | NobodyCam, yes today meeting i saw it | 19:31 |
| NobodyCam | :) | 19:31 |
| Nisha | NobodyCam, so should i base my code changes on those code changes right now? or it will be better to rebase them once devananda's work is merged | 19:33 |
| *** dprince has quit IRC | 19:34 | |
| NobodyCam | oh thats a tuff call I have only done a quick look over. I would say base the work off the devananda's but I'm really sure how close it landing it is. | 19:35 |
| NobodyCam | *...reallly NOT sure... | 19:35 |
| Nisha | :) | 19:36 |
| Nisha | NobodyCam, Ok so i will put the code changes for review independently for now and when states code is merged will rebase my changes on top of that | 19:36 |
| NobodyCam | sound good... please note that in a comment or the commit message | 19:37 |
| *** marcoemorais has quit IRC | 19:39 | |
| *** marcoemorais has joined #openstack-ironic | 19:40 | |
| Shrews | devananda: Is there a reason to move the helper methods out of the ConductorManager class in 139217? | 19:40 |
| *** marcoemorais has quit IRC | 19:40 | |
| devananda | Shrews: yes. I just haven't gotten around to writing it yet | 19:40 |
| devananda | Shrews: see aweeks' comment on the first patch | 19:41 |
| *** marcoemorais has joined #openstack-ironic | 19:41 | |
| *** marcoemorais has quit IRC | 19:41 | |
| *** marcoemorais has joined #openstack-ironic | 19:41 | |
| Shrews | devananda: i see no comments on any patch | 19:41 |
| devananda | NobodyCam: I think jroll would like us to land my fsm changes asap | 19:41 |
| NobodyCam | devananda: I just started to take a closer look at it now | 19:42 |
| NobodyCam | :) | 19:42 |
| Shrews | devananda: oh, you mean the first review | 19:42 |
| devananda | Shrews: L121 here https://review.openstack.org/#/c/139215/1/ironic/common/fsm.py | 19:42 |
| Shrews | yup | 19:42 |
| *** Marga_ has joined #openstack-ironic | 19:44 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Refactoring: split test.py and rewrite tests for process https://review.openstack.org/140339 | 19:47 |
| *** penick has quit IRC | 19:50 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Refactoring: split test.py and rewrite tests for process https://review.openstack.org/140339 | 19:51 |
| *** marcoemorais has quit IRC | 19:52 | |
| *** rushiagr_away is now known as rushiagr | 19:52 | |
| *** marcoemorais has joined #openstack-ironic | 19:52 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-python-agent: Use LLDP to get switch port mapping https://review.openstack.org/92627 | 19:59 |
| JoshNang | JayF: jroll ^ | 19:59 |
| JoshNang | rebased | 19:59 |
| JayF | NobodyCam: had a vote on that too ^^ | 19:59 |
| NobodyCam | huh | 20:00 |
| NobodyCam | oh | 20:00 |
| NobodyCam | hehehe | 20:00 |
| NobodyCam | JoshNang: Thank you for the additional logging | 20:01 |
| adam_g | JayF, hmm. that devstack fix seems not to have solved the problem | 20:02 |
| JayF | adam_g: Are you sure the failure you're looking at had you change in it? | 20:03 |
| JayF | I think with devstack changes they have to wait for nodes to cycle and rebuild for it to take effect or smth like that | 20:03 |
| *** anderbubble has joined #openstack-ironic | 20:03 | |
| adam_g | JayF, yup. ssh key creation happens a full 3 minutes prior to nodes enrolling | 20:03 |
| JayF | :( | 20:03 |
| *** romcheg1 has joined #openstack-ironic | 20:05 | |
| Shrews | devananda: to speed things along, would you mind if i pushed a patch to your fsm change to remove the py26 import and remove comments with manager.py:<LINE> in states.py? | 20:05 |
| devananda | Shrews: not at all | 20:05 |
| * Shrews does | 20:06 | |
| devananda | Shrews: i'm distracted with other things right now, feel free to update that series | 20:06 |
| JoshNang | NobodyCam: no problem! thanks for catching that. definitely saved someone some debugging problems down the road | 20:06 |
| *** spandhe has quit IRC | 20:07 | |
| *** Marga_ has quit IRC | 20:09 | |
| *** Marga_ has joined #openstack-ironic | 20:09 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Discover node properties using new CLI node-inspect https://review.openstack.org/100951 | 20:10 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Discover node properties for iLO drivers https://review.openstack.org/103007 | 20:13 |
| *** jrist has quit IRC | 20:14 | |
| *** spandhe has joined #openstack-ironic | 20:16 | |
| rloo | Shrews: wrt the fsm and the conductor line numbers. Maybe replace with the method name for now? | 20:17 |
| Shrews | rloo: i still do not like that, but if others want it, i can change it to method name | 20:19 |
| *** Marga_ has quit IRC | 20:20 | |
| rloo | Shrews: I think it is useful for reviewing this code. I know what you mean, but am thinking (maybe) that the method names won't change before we transition to the new state machine. | 20:20 |
| NobodyCam | Shrews: are you adding the logging stuff Josh H pointed out in states.py? | 20:20 |
| *** andreykurilin_ has quit IRC | 20:20 | |
| rloo | Shrews: we can have a low hanging fruit for someone to delete those comments after the series has merged ;) | 20:20 |
| NobodyCam | worth me adding a +1 to the current review? | 20:21 |
| Shrews | NobodyCam: i can, if my vm ever unfreezes | 20:22 |
| NobodyCam | lol :( | 20:23 |
| Shrews | which isn't looking promising, atm :( | 20:23 |
| rloo | Shrews: just get a new cow ;) | 20:23 |
| NobodyCam | rloo: http://www.wikihow.com/Have-a-Pet-Cow | 20:24 |
| NobodyCam | lol | 20:24 |
| rloo | NobodyCam: :D | 20:24 |
| Shrews | ugh | 20:25 |
| * Shrews reboots | 20:25 | |
| Shrews | brb | 20:25 |
| *** jrist has joined #openstack-ironic | 20:26 | |
| *** rushiagr is now known as rushiagr_away | 20:27 | |
| *** marcoemorais has quit IRC | 20:34 | |
| *** marcoemorais has joined #openstack-ironic | 20:34 | |
| *** marcoemorais has quit IRC | 20:35 | |
| *** marcoemorais has joined #openstack-ironic | 20:41 | |
| Shrews | rloo: i can address some of your comments myself in what i'm about to push, but not all of them. i didn't write the code, and i'm pretty sure deva just copied most of it, so he may not have the answers either :) | 20:41 |
| rloo | Shrews -- address as many as you can. I can ping harlowja about the others. thx. | 20:42 |
| *** anderbubble has quit IRC | 20:42 | |
| Shrews | k k | 20:42 |
| rloo | Shrews: I'm going to try to play around with the fsm stuff. some of my questions could be cuz I don't have a clear idea yet of how it works. | 20:42 |
| rloo | Shrews: Am I right that "we" all want this fsm stuff? (cuz there's no spec for this) | 20:43 |
| Shrews | i think the exception strings should definitely be _()-ized | 20:43 |
| NobodyCam | rloo: there is a spec.. this is to enable the new state machine | 20:44 |
| Shrews | yeah, that | 20:44 |
| rloo | NobodyCam: ha ha. well, that spec says there are other specs with implementation details ;) | 20:44 |
| rloo | NobodyCam: I just want to make sure we're following our own rules ;) | 20:45 |
| jroll | rloo: this is about expressing the current state machine in a sane fashion, so we can more easily evaluate the changes needed to get to the new one | 20:45 |
| NobodyCam | we could file a "Ironic's current state machine lacks code" bug | 20:46 |
| devananda | rloo: it isn't tagged as 'implements a blueprint' -- but yues, this is all related to the state machine spec | 20:46 |
| devananda | rloo: I can tag it, if that will help with tracking / sanity | 20:46 |
| NobodyCam | I would reather not mix it in with the NEW state machine stuff | 20:47 |
| devananda | rloo: also, shrews is correct - i copied fsm.py from TaskFlow. you can see the full history of my changes (which I squashed before posting to gerrit) on my github, if you're interested | 20:47 |
| devananda | NobodyCam: right -- and this is all prerequisite | 20:47 |
| rloo | yeah, I realize why we're adding the fsm, and I'm fine with it. but it is a fairly substantial change and there's no spec or anything that mentions this new feature. | 20:48 |
| devananda | rloo: do we actually need one? or is that just process for process' sake? | 20:48 |
| NobodyCam | I was half serious about the bug thing | 20:48 |
| rloo | devananda: I don't know if we need one. But we do ask others to write up specs, so I thought I'd ask. What do people think? | 20:48 |
| rloo | devananda: maybe we can just update the existing state machine spec to mention using taskflow to represent the FSM etc, and that might be good enough? | 20:49 |
| devananda | rloo: how about a "related-to: BP nnnnn" line in the commit messages? | 20:49 |
| NobodyCam | that would work for me | 20:50 |
| devananda | rloo: hm. I dont think victor_lowther's spec needs to indicate exactly how the state machine is implemented, whether its borrowed from taskflow, or what ever | 20:50 |
| *** anderbubble has joined #openstack-ironic | 20:50 | |
| rloo | devananda: it doesn't. but don't you think this state machine is a feature, not a bug? | 20:50 |
| devananda | it does indicate that we need to have _A_ representation of a state machine | 20:50 |
| devananda | and here's code to create one -- so yea, the commits should reference the spec | 20:50 |
| devananda | i'm convinved now :) | 20:50 |
| NobodyCam | lol .... /me makes a fresh pot of coffee | 20:51 |
| rloo | devananda: ;) | 20:51 |
| *** mrda-away is now known as mrda | 20:51 | |
| mrda | Morning Ironic | 20:51 |
| *** bradjones has quit IRC | 20:52 | |
| rloo | mornin' mrda | 20:54 |
| NobodyCam | morning mrda | 20:55 |
| *** wanyen has quit IRC | 20:56 | |
| mrda | o/ | 20:57 |
| *** Nisha has quit IRC | 20:57 | |
| *** dprince has joined #openstack-ironic | 20:57 | |
| *** marcoemorais has quit IRC | 20:57 | |
| *** subscope has joined #openstack-ironic | 20:57 | |
| rloo | hey harlowja, wrt your comments in https://review.openstack.org/#/c/139215. eg, 'Another change u probably want to do that I need to get a fix for...' | 21:01 |
| rloo | harlowja: does that mean you'll update this code? | 21:01 |
| NobodyCam | devananda: would a shallow copy work here? https://review.openstack.org/#/c/139216/3/ironic/conductor/task_manager.py Line#184 | 21:01 |
| Shrews | devananda: i'll add the related-to for you | 21:03 |
| *** harlowja is now known as harlowja_away | 21:04 | |
| JayF | oooh | 21:04 |
| JayF | sounds like I can pull ordereddict from IPA soon too :) | 21:05 |
| adam_g | JayF, doh, i think this is the real issue/PITA: https://bugs.launchpad.net/ironic/+bug/1400882 | 21:06 |
| adam_g | not sure thats a bug but created a new one from the gate issue, in case | 21:07 |
| JayF | I mean, seems buggyish to me? | 21:07 |
| JayF | the new enrollment stuff laid out in the state machine will help | 21:07 |
| JayF | but it seems like for now you just need a busy loop saying Devstack isn't ready to move on until those nodes are in the resource tracker? | 21:08 |
| jroll | no it won't, because you'll still have to wait for nova | 21:08 |
| JayF | nova is the only thing with a resource tracker, right? | 21:09 |
| PaulCzar | I'm up to my /last/ problem with ironic ... which is that the instance can't resolve DNS while in the bootloader stage, which means my instances can't access ironic api to tell ironic to do the iscsi stuff | 21:09 |
| JayF | if you're using neutron for dhcp | 21:10 |
| PaulCzar | setting endpoint in keystone to IP makes it work ... but then I lose not only DNS .. but SSL | 21:10 |
| adam_g | JayF, yeah, polling nova's view of memory/etc as well as node count. | 21:10 |
| JayF | you probably just need to have it ship along dns server ips as well | 21:10 |
| openstackgerrit | David Shrewsbury proposed openstack/ironic: Add fsm.py module https://review.openstack.org/139215 | 21:10 |
| PaulCzar | JayF: it does! if I spoof the MAC on a VM running ubuntu it comes up with dns working | 21:10 |
| openstackgerrit | Merged openstack/ironic-python-agent: Use LLDP to get switch port mapping https://review.openstack.org/92627 | 21:11 |
| *** anderbubble_ has joined #openstack-ironic | 21:11 | |
| NobodyCam | PaulCzar: have you looked at https://ask.openstack.org/en/question/26278/neutron-dhcp-adding-dns/ | 21:11 |
| JayF | PaulCzar: strange. IDK anything beyond that then, because I don't use the pxe ramdisk. I could help if it was IPA :) | 21:11 |
| *** anderbubble has quit IRC | 21:11 | |
| *** anderbubble_ is now known as anderbubble | 21:11 | |
| *** harlowja_away is now known as harlowja | 21:12 | |
| Shrews | NobodyCam: doh! forgot the logging stuff | 21:12 |
| NobodyCam | :-p | 21:13 |
| harlowja | rloo as for 139215 i'm waiting to see where this goes first, before thinking about how to generalize it, i think thats fair with regard to the FSM stuff | 21:13 |
| harlowja | i have done something @ https://github.com/harlowja/automaton that might fit the needs for both projects, but it seems a little premature to start adding it to the requirements repo and such so far... | 21:14 |
| Shrews | NobodyCam: where does it go?? | 21:14 |
| rloo | harlowja: yes, that's fair. so we should just go ahead and make any changes directly to our copy? | 21:14 |
| harlowja | ya | 21:14 |
| rloo | harlowja: ok | 21:14 |
| harlowja | doesn't seem like that review's even merged yet, so not sure the point until the 'vision is more clear' | 21:15 |
| PaulCzar | JayF: IPA requires that there's already an OS laid down on the box running the IPA right ? | 21:15 |
| *** Masahiro has joined #openstack-ironic | 21:15 | |
| harlowja | if vision gets more clear then sure rloo we can decide what to do | 21:15 |
| rloo | harlowja: merging == vision is crystal clear :-) | 21:15 |
| harlowja | lol | 21:15 |
| JayF | PaulCzar: IPA is just another deploy driver. Right now they support different things (IPA only does whole disk images, and requires images from glance backed by swift) | 21:15 |
| harlowja | thats impossible rloo | 21:15 |
| rloo | harlowja: ha ha | 21:16 |
| JayF | PaulCzar: IPA runs inside a ramdisk just like the pxe driver stuff does; it just offloads more work to the node. i.e. IPA has a rest API that's called with a URL to an image, the agent itself downloads and images it, as opposed to pxe driver, where it mounts an iscsi volume and then lays an image down from the conductor | 21:16 |
| harlowja | rloo unless u and ironic folks can predict the future (if so please share some numbers with me, ha) | 21:16 |
| rloo | harlowja: "crystal clear" doesn't say anything about the future ;) | 21:17 |
| Shrews | NobodyCam: hrm, that's not actually a small change. i'm not comfortable adding that logging stuff yet | 21:17 |
| harlowja | rloo fair enough :) | 21:17 |
| NobodyCam | Shrews: ack I was just looking at that | 21:18 |
| harlowja | although rloo if u want to go through a fine toothed comb u can go through https://github.com/harlowja/automaton and submit PRs ;) | 21:18 |
| Shrews | NobodyCam: if i knew the code better... | 21:18 |
| Shrews | so maybe later | 21:18 |
| NobodyCam | :) | 21:18 |
| *** Masahiro has quit IRC | 21:19 | |
| rloo | harlowja: what's a PR? I meant making sure our FSM reflected what our code is/was doing. | 21:19 |
| *** cohn has left #openstack-ironic | 21:20 | |
| harlowja | rloo pull request (that weird github thing) | 21:20 |
| *** spandhe has quit IRC | 21:20 | |
| rloo | harlowja: ahh. so what's the diff between harlowja/automaton and your TaskFlow stuff? | 21:21 |
| harlowja | a few diffs | 21:21 |
| harlowja | nothing major | 21:21 |
| harlowja | moved the runners to be external to the state machine they run using | 21:21 |
| harlowja | and put in hierachical state machines into that one (for those that need it) | 21:22 |
| harlowja | otherwise, the same | 21:22 |
| harlowja | * https://github.com/harlowja/automaton/blob/master/automaton/machines.py#L370 | 21:22 |
| *** marcoemorais has joined #openstack-ironic | 21:23 | |
| rloo | harlowja: I'm worried about maintaining our copy of your fsm, but maybe I'll postpone the worrying and hopefully forget to worry about it. | 21:23 |
| PaulCzar | JayF: is there an example somewhere using IPA? it's not on the deploying ironic documentation | 21:23 |
| harlowja | rloo i think the worry is justified, but need to see where this goes first | 21:23 |
| harlowja | if it goes no-where, then no need to worry | 21:23 |
| rloo | harlowja: ok, i'll punt on the worrying then ;) | 21:23 |
| harlowja | :) | 21:23 |
| JayF | PaulCzar: I think it is in the documentation? I can't really spare the time right now to help one-on-one although I could later (tomorrow) | 21:24 |
| rloo | harlowja: we're going to use it... | 21:24 |
| harlowja | rloo thats decided already? | 21:24 |
| PaulCzar | JayF: Okay I'll poke around and see what I can find for myself | 21:24 |
| JayF | PaulCzar: or ask others in here :/ | 21:24 |
| harlowja | rloo if so then maybe the worrying is justified, ha | 21:24 |
| rloo | harlowja: we just need 2 +2s and an approve ;) | 21:24 |
| harlowja | well i have put up https://pypi.python.org/pypi/automaton and can iterate on that as needed, so if so it'd be nice to try to unify on a common library | 21:25 |
| harlowja | rloo maybe u want to be the owner of that library :-P | 21:25 |
| rloo | harlowja: no no no. i'm not worrying about that stuff, remember? | 21:25 |
| harlowja | oh ya | 21:25 |
| harlowja | :) | 21:25 |
| ekarlso- | rloo: isonic is using a copy of automaton vs using the lib ? | 21:26 |
| rloo | ekarlso-: well, ironic isn't using anything yet, but yeah, we're talking about this patch: https://review.openstack.org/#/c/139215 | 21:27 |
| harlowja | ekarlso- i created automaton over the weekend, so its sorta hard for people to use it yet ;) | 21:27 |
| harlowja | i'm trying to be pre-emptive :-P | 21:27 |
| harlowja | *not releated to preemptive nuclear strikes | 21:27 |
| harlowja | lol | 21:27 |
| PaulCzar | Anyone have some time to talk me through using IPA or able to show me to some docs ? | 21:28 |
| harlowja | ekarlso- automaton comes from taskflow (which uses it internally to itself) and if others find something similar useful, might as well share the love/code | 21:29 |
| harlowja | * basically http://docs.openstack.org/developer/taskflow/types.html#module-taskflow.types.fsm | 21:32 |
| JoshNang | PaulCzar: sure! what questions do you have? | 21:32 |
| JoshNang | i'm not sure how many docs we have tbh | 21:32 |
| *** marcoemorais has quit IRC | 21:33 | |
| *** marcoemorais has joined #openstack-ironic | 21:33 | |
| PaulCzar | JoshNang: right now I'm using the pxe_ssh driver which is causing me grief because the initial bootloader doesn't seem to do DNS ... so I'm trying to figure out of there's a better or alternative driver that might help | 21:34 |
| *** alexpilotti has quit IRC | 21:35 | |
| PaulCzar | JoshNang: a link to a prebuild IPA loader and some quick notes on the node options to set ( equivs of -i pxe_deploy_kernel=... etc ) would probably be enough to get me in the right direction | 21:36 |
| JoshNang | PaulCzar: hrm, i'm not sure ipa is going to help if your ramdisk isn't getting DNS. IPA will need dns to find glance :/ | 21:36 |
| PaulCzar | JoshNang: neutron uses the same lease for the ramdisk as it does for the actual built image right ? | 21:38 |
| *** linggao has quit IRC | 21:38 | |
| JoshNang | PaulCzar: hmm, i think so. let me check | 21:39 |
| PaulCzar | JoshNang: thanks. because dns works fine one the instance is up ... so I've been assuming its a problem with the bootloader from DIB itself | 21:40 |
| *** Marga_ has joined #openstack-ironic | 21:40 | |
| *** subscope has quit IRC | 21:42 | |
| JoshNang | PaulCzar: yeah i'm not seeing any code that would change it's lease | 21:42 |
| JoshNang | but i'm far from an expert on how the pxe driver works | 21:43 |
| PaulCzar | I assume with the IPA driver it still uses pxe to start the bootloader that contains the IPA ? | 21:44 |
| JoshNang | correct | 21:44 |
| JoshNang | here's some extremely barebones docs (including download link) https://github.com/openstack/ironic-python-agent/tree/master/imagebuild/coreos | 21:45 |
| NobodyCam | brb | 21:45 |
| JoshNang | for us, we use pxe to boot into ipxe, and then ipxe downloads the kernel/initrd and boots them | 21:46 |
| PaulCzar | JoshNang: and that pxe is outside of ironic itself ? has nothing to do with the pxe driver ? | 21:50 |
| JoshNang | PaulCzar: correct. "pxe driver" is a bit confusing, as they both pxe | 21:51 |
| JoshNang | "pxe driver" == "write image via iSCSI using a pxe booted ramdisk" | 21:51 |
| JoshNang | "ipa" == "write image directly to disk after downloading from URL to pxe booted ramdisk" | 21:52 |
| devananda | can we rename the pxe_* drivers to iscsi_* pls? | 21:52 |
| devananda | without, you know, breaking the world | 21:52 |
| JoshNang | +100 | 21:52 |
| PaulCzar | JoshNang: right ... also the pxe driver handles the heavy lifting of managing your images/macs in /tftpboot | 21:52 |
| devananda | the whole split-boot-and-deploy-interfaces seems to have been dropped lately, which is unfortunate | 21:52 |
| devananda | or maybe I jsut haven't been following it? | 21:53 |
| JoshNang | PaulCzar: the agent driver uses the same code to handle the images/macs in tftpboot | 21:53 |
| *** jrist is now known as knowncitizen | 21:54 | |
| *** knowncitizen is now known as jrist | 21:55 | |
| NobodyCam | omg 3 +2's | 21:59 |
| NobodyCam | lol | 21:59 |
| *** harlowja has quit IRC | 22:02 | |
| PaulCzar | JoshNang: what are the -i options for the agent driver to specify the kernel/initrd ? | 22:05 |
| *** ryanpetrello has quit IRC | 22:05 | |
| *** foexle has quit IRC | 22:07 | |
| JoshNang | PaulCzar: should be the same as the pxe driver (and if not, i'd say that's a bug) | 22:07 |
| *** andreykurilin_ has joined #openstack-ironic | 22:09 | |
| rloo | devananda: i was going to ask about boot/deploy interfaces, cuz you wanted it for kilo1. lucas mentioned that he's too busy to work on it, and jroll offered to try to pick it up. | 22:10 |
| devananda | I jhave the sense jroll is too busy, too ... | 22:10 |
| *** spandhe has joined #openstack-ironic | 22:12 | |
| *** sdake has joined #openstack-ironic | 22:12 | |
| rloo | devananda: what's the priority on the boot/deploy? seems like the new state stuff is higher priority? | 22:12 |
| sdake | http://sdake.io/2014/12/09/isnt-it-atomic-on-openstack-ironic-dont-you-think/ | 22:12 |
| sdake | enjoy ^^ ;-) | 22:12 |
| *** alexpilotti has joined #openstack-ironic | 22:13 | |
| rloo | sdake, have you thought about updating/adding to the ironic docn? :-) | 22:14 |
| PaulCzar | JoshNang: and finally ... the image format of the images that it supports? It was mentioned earlier that it does whole disk ... so would that be a qcow or a raw image? | 22:14 |
| *** hypnosb has joined #openstack-ironic | 22:14 | |
| PaulCzar | or is it still the set of three image+kernel+initrd | 22:15 |
| JoshNang | i believe qcow's | 22:16 |
| sdake | what is a docn? | 22:16 |
| rloo | sdake: docn -- documentation. or lack of ;) | 22:16 |
| sdake | ironic haas good docks | 22:16 |
| sdake | docs | 22:16 |
| sdake | so my thinknig is I'd like to get that blueprint impolemented | 22:17 |
| sdake | so atomic really works on ironic without alot of hassle | 22:17 |
| JayF | PaulCzar: anything that qemu-img convert can convert to raw :) | 22:17 |
| sdake | (blueprint at the end) | 22:17 |
| JayF | PaulCzar: are supported image types | 22:17 |
| JayF | lol | 22:17 |
| JoshNang | JayF: thanks :) | 22:17 |
| hypnosb | Hi | 22:17 |
| PaulCzar | JayF: thanks :) | 22:18 |
| hypnosb | So question... Does it work? | 22:18 |
| devananda | sdake: iiuc, this is just in VMs -- did you get devstack deploying Atomic on separate hardware? | 22:18 |
| devananda | hypnosb: that's a strange question | 22:18 |
| rloo | sdake: do you know about our BP spec process? | 22:18 |
| hypnosb | devananda, Asking a software project if their project works is strange? | 22:19 |
| devananda | hypnosb: define "it" and "works" :) | 22:19 |
| hypnosb | Does it do what you say it does? | 22:19 |
| devananda | hypnosb: yup | 22:19 |
| sdake | rloo no I do not | 22:20 |
| sdake | if you ahve a pointer | 22:20 |
| devananda | hypnosb: but that's not "does it work" -- that's "are your docs accurate" | 22:20 |
| hypnosb | amazing | 22:20 |
| sdake | devandanda I'd assume it would work on bare metal as well | 22:20 |
| hypnosb | id rather have accurate docs than having it work... id go less crazy | 22:20 |
| sdake | rloo if you have a pointer I'll follow the process therre | 22:20 |
| sdake | or an example review | 22:20 |
| devananda | sdake: I would assume so, too. but it's always interesting to try :) | 22:20 |
| rloo | sdake: https://github.com/openstack/ironic-specs | 22:20 |
| devananda | sdake: http://specs.openstack.org/openstack/ironic-specs/specs/template.html | 22:21 |
| sdake | rloo cool I'll fiel that | 22:21 |
| devananda | sdake: https://wiki.openstack.org/wiki/Blueprints#Spec_.2B_Blueprints_lifecycle | 22:21 |
| hypnosb | how is ironic different from fuel? | 22:21 |
| devananda | hypnosb: ehhh... very different .... | 22:21 |
| devananda | hypnosb: but also, not that different | 22:21 |
| devananda | much too long / loaded a question to ask me right now, as I've just spent the last four hours replyng to emails about FuelAgent | 22:21 |
| rloo | sdake: so you file a blueprint (you did that), and then write up a specification. Otherwise that BP may sit there gathering dust. | 22:22 |
| sdake | rloo will do | 22:22 |
| devananda | hypnosb: have you read the docs for each project? it should then be clear :) | 22:22 |
| sdake | rloo thanks for the hand holding ;-) | 22:22 |
| rloo | sdake: yw | 22:23 |
| Shrews | sdake: without me having to read the entire blog post (i'm lazy and it's almost beer-o-clock), what role is docker playing? | 22:23 |
| hypnosb | devananda, ahahah nice one | 22:23 |
| devananda | sdake: these are *terribly* out of date, but, are still actually pretty close to graphics covering the process you didn't find | 22:23 |
| devananda | sdake: https://github.com/openstack/tripleo-incubator/tree/master/images | 22:23 |
| sdake | shrews docker does nothing - it is about launching atomic on ironic | 22:24 |
| sdake | but I am working on magnum which controls docker | 22:24 |
| sdake | magnum uses ironic to launch bare metal | 22:24 |
| sdake | the bare metal image needs docker (a dependency of magnum) | 22:24 |
| hypnosb | does ironic use puppet or chef? | 22:24 |
| Shrews | sdake: ah, ok. assumed it played a role from the 1st paragraph | 22:24 |
| sdake | its just an explination of "why" you would even want atomic ;-) | 22:24 |
| sdake | "who needs more operating systems." | 22:25 |
| Shrews | i should just RTFBP :) | 22:25 |
| sdake | its mostly documenting my steps for others | 22:25 |
| sdake | but I reproduced it after I hacked together an initial implementation | 22:25 |
| sdake | so it should reproduce for others as well | 22:25 |
| hypnosb | I thought crowbar was taking off but I guess the dell guys killed it | 22:25 |
| sdake | AFAIK ironic does not use puppet or chef, but you could use heat software config resources to do that job | 22:26 |
| sdake | btw, +1 from me on ironic architecture and implementation | 22:26 |
| sdake | having been through all of it, it looks like pretty hot tech | 22:26 |
| hypnosb | my thought for puppet or chef was the bare metal provision... I guess its all bash script? | 22:26 |
| hypnosb | python | 22:27 |
| sdake | hypnosb provision can be done with cloudinit using heat software config | 22:27 |
| devananda | whoa, conflation of concepts here folks | 22:27 |
| hypnosb | sure but what about bare metal of a node? | 22:27 |
| devananda | sdake: you mean s/provision/parameterization/ | 22:28 |
| hypnosb | oh | 22:28 |
| hypnosb | er | 22:28 |
| hypnosb | yeah idk how heat can do bare metal? | 22:28 |
| devananda | sdake: parameterization of an instance, after it has been deployed, can be done by chef/puppet/salt/ansible/anything | 22:28 |
| devananda | cloud-init + os-apply-config can do that parameterization as well | 22:28 |
| devananda | hypnosb: look at tripleo | 22:28 |
| devananda | heat drives nova, and passes metadata through cloud-init | 22:29 |
| hypnosb | omg another openstack project | 22:29 |
| hypnosb | ffs | 22:29 |
| devananda | os-*-config scripts run within the instance at boot, gather the metadata, and parameterize | 22:29 |
| devananda | hypnosb: hi. what's with all the hate? | 22:29 |
| hypnosb | frustrated, 5:30pm | 22:30 |
| hypnosb | ill go | 22:30 |
| hypnosb | sorry, bye | 22:30 |
| devananda | hypnosb: your first words in this channel were not friendly. perhaps try again tomorrow :) | 22:30 |
| sdake | devandanda yup I thnk I siaid that, just differently ;-) | 22:31 |
| devananda | sdake: possibly :) the part that confused me was "provision can be done with cloudinit". | 22:32 |
| alexm__ | Hi everyone, wondering what you are using to build your bootable metal images for Ironic and if we need to do anything special to make them compatible. My ironic deployment with nova succeeds, but the node doesn’t want to boot on the written disk partition :( | 22:33 |
| devananda | sdake: at least in the context of ironic, provisioning means something specific, which doesn't have anything to do with cloud-init | 22:33 |
| *** dprince has quit IRC | 22:33 | |
| devananda | alexm__: hi! we're using diskimage-builder, but in principle, images built by other tools should work too | 22:34 |
| NobodyCam | Shrews: I think I got that logging stuff mind if I push up over 139215 | 22:34 |
| NobodyCam | devananda: also ^^^ | 22:34 |
| Shrews | NobodyCam: YES! | 22:34 |
| PaulCzar | JoshNang: looks like pxe driver uses pxe_deploy_kernel and agent uses deploy_kernel | 22:34 |
| NobodyCam | yes you mind? | 22:34 |
| Shrews | NobodyCam: i mean... NO! | 22:34 |
| Shrews | :) | 22:34 |
| NobodyCam | lol | 22:34 |
| NobodyCam | hehehe | 22:34 |
| Shrews | NobodyCam: please do it | 22:34 |
| devananda | sdake: nice blog post, btw | 22:35 |
| alexm__ | devananda: thx. I used diskimage-builder to make my ubuntu and kernel+ramdisk images. Unfortunately it doesn’t seem to boot. | 22:35 |
| devananda | sdake: if you feel like diagraming, I'd love to have the deploy process for the pxe_* class of drivers, done using asciiflow, in our doc/source | 22:36 |
| JoshNang | PaulCzar: :/ | 22:36 |
| devananda | sdake: I have this terribly old one here ... https://github.com/openstack/tripleo-incubator/blob/master/images/baremetal-deploy.odp which I haven't updated in years :( | 22:36 |
| openstackgerrit | Chris Krelle proposed openstack/ironic: Add fsm.py module https://review.openstack.org/139215 | 22:36 |
| NobodyCam | Shrews: ^^^ I think thats kinda it | 22:36 |
| devananda | alexm__: ok. I'd need more information to be able to help determine what the problem is | 22:36 |
| *** ryanpetrello has joined #openstack-ironic | 22:37 | |
| *** harlowja has joined #openstack-ironic | 22:37 | |
| devananda | NobodyCam: let's remove run_iter() -- I dont think we'll ever use it (famous last words ...) | 22:39 |
| sdake | devananda I dont know asciflow, and at the moment don't have time until January to dig into any new work | 22:39 |
| alexm__ | devananda: All the service logs look okay, from the point of view of conductor and nova it was a success. Maybe there was a problem witth the pxe config. Do you have a valid one that I can compare with? | 22:39 |
| harlowja | devananda in https://github.com/harlowja/automaton i split off the runner to be its own class, might help for those who don't care about using it | 22:39 |
| sdake | devananda i'd be more willing if I could use imagemagick files | 22:39 |
| harlowja | *including run_iter() | 22:40 |
| sdake | (eg SVG) | 22:40 |
| devananda | sdake: either this https://www.websequencediagrams.com/# | 22:40 |
| devananda | sdake: or this: http://asciiflow.com/ | 22:40 |
| sdake | is asciiflow integrated into sphinx in some way? | 22:41 |
| devananda | both are text representations that can be embedded in REST files, and later edited, and which generate pictures | 22:41 |
| devananda | nope | 22:41 |
| devananda | but the output can easily be pasted into a .rst file | 22:41 |
| sdake | I see | 22:41 |
| sdake | interesting | 22:41 |
| devananda | see eg. https://review.openstack.org/#/c/133828/12/specs/kilo/new-ironic-state-machine.rst | 22:41 |
| sdake | I'll look into it after milestone #1 of magnum is out | 22:41 |
| devananda | for several complex diagrams in asciiflow | 22:42 |
| devananda | sdake: ack. anyway, thanks! | 22:42 |
| *** krtaylor has quit IRC | 22:42 | |
| sdake | sure, hope the blog post attracts some more attention to ironic (and atomic) | 22:42 |
| *** ryanpetrello has quit IRC | 22:44 | |
| NobodyCam | harlowja: question: can I ask why this check: https://github.com/openstack/taskflow/blob/master/taskflow/engines/action_engine/runner.py#L166 | 22:45 |
| harlowja | u may ask :-P | 22:45 |
| NobodyCam | :) | 22:45 |
| harlowja | NobodyCam meh, just avoids adding on un-needed things, nothing much else there | 22:45 |
| harlowja | avoid any callbacks if no debuggin on | 22:45 |
| harlowja | save some cycles, blah blah :-P | 22:46 |
| *** ryanpetrello has joined #openstack-ironic | 22:46 | |
| harlowja | i know its python... | 22:46 |
| harlowja | ha | 22:46 |
| NobodyCam | ok thats what I thought | 22:46 |
| NobodyCam | :) | 22:46 |
| NobodyCam | Thank you | 22:46 |
| harlowja | np | 22:46 |
| PaulCzar | JoshNang: the bootloader started ... but it crashed out to a prompt at 'Failed to start Switch Root' | 22:47 |
| * devananda steps away for a break | 22:47 | |
| JoshNang | PaulCzar: weird! | 22:48 |
| PaulCzar | JoshNang: that's with the prebuilt images I downloaded ... digging into the problem there appear to be some files missing in the image so I'll create images locally and try again | 22:49 |
| JoshNang | PaulCzar: hmmm i'm pretty sure we're using those images in the gate right now. what files are missing? | 22:49 |
| JayF | PaulCzar: not all ubuntu cloud images you can download come with hw drivers :) | 22:50 |
| NobodyCam | d'oh | 22:50 |
| NobodyCam | Shrews: are you pushing up another 139215? | 22:50 |
| PaulCzar | JayF: this is still at the bootloader stage ... its a coreos error | 22:51 |
| Shrews | No. I'm pushing a beer into my face hole and making dinner | 22:51 |
| JayF | PaulCzar: is that an error from booting IPA then? | 22:51 |
| PaulCzar | yeah | 22:51 |
| NobodyCam | lol.. ack I will fix my pep8 errors :-p | 22:51 |
| *** mjturek has quit IRC | 22:52 | |
| JayF | PaulCzar: try redownloading it and seeing if the md5 matches? | 22:52 |
| JayF | PaulCzar: there is a small race where new IPA images aren't uploaded atomically, and I bet that could've been what happened to you | 22:52 |
| JayF | PaulCzar: (i.e., if you download while we're uploading you can get a bad image) | 22:52 |
| JayF | PaulCzar: also, alternatively, build your own. It's trivial to do. | 22:52 |
| PaulCzar | yeah that's what I'm doing | 22:53 |
| openstackgerrit | Chris Krelle proposed openstack/ironic: Add fsm.py module https://review.openstack.org/139215 | 22:53 |
| PaulCzar | md5 checksums match ... I'll see what happens on a self built set | 22:57 |
| *** jgrimm is now known as zz_jgrimm | 22:59 | |
| PaulCzar | I assume the images are supposd to go into glance with disk-format aki/ari ? | 22:59 |
| *** bigjools has quit IRC | 23:00 | |
| sdake | ramdisk goes in as ari, kernel goes in as aki | 23:00 |
| sdake | base image goes in as ami | 23:00 |
| sdake | the base image must be a filesystem (use the file command to double check) | 23:01 |
| sdake | the base image must be registered with ramdisk_id and kernel_id glance properties | 23:01 |
| PaulCzar | I'm using the same base image that I used for the pxe_ssh driver | 23:01 |
| *** bigjools has joined #openstack-ironic | 23:02 | |
| *** bigjools has quit IRC | 23:02 | |
| *** bigjools has joined #openstack-ironic | 23:02 | |
| *** alexpilotti has quit IRC | 23:02 | |
| sdake | did you glance image-create --property kernel_id=xyz --proeprty ramdisk_id=xyz ? | 23:02 |
| PaulCzar | yessir | 23:02 |
| sdake | i'm out of ideas, but keep in mind I've only been using ironic for about 3 days :) | 23:02 |
| *** Masahiro has joined #openstack-ironic | 23:04 | |
| PaulCzar | sdake: as I've been discovering over the last few weeks, it all works great in devstack ... just trying to run it in a real openstack is 'fun' | 23:04 |
| sdake | I see, I haven't actually tried real hardware | 23:04 |
| sdake | I only have one machine I can use, and I don't want to lose the disk on it | 23:04 |
| *** shakamunyi has joined #openstack-ironic | 23:04 | |
| sdake | rather two machines, I have my workststation where I could run a conductor and a machine that I could deploy on | 23:04 |
| PaulCzar | yeah, also devstack hides a ton of complexity in impossible to read shell scripts all over the place | 23:05 |
| sdake | there are installation documentation that looked pretty good I think | 23:05 |
| sdake | are you sure your root image is correct? | 23:05 |
| sdake | you can check by running file on it | 23:05 |
| sdake | it should print out that its ext3/ext4/etc | 23:05 |
| *** Masahiro has quit IRC | 23:08 | |
| PaulCzar | sdake: yeah my problems aren't with the image itself ... its with the deployers | 23:10 |
| sdake | what driver are you using? | 23:10 |
| PaulCzar | the pxe_ssh deploy images don't seem to get dns info | 23:10 |
| PaulCzar | and I just remembered why I wasn't going to use the agent driver ... it requries swift | 23:10 |
| PaulCzar | which is a lot of overhead for provisioning machines | 23:11 |
| *** mikedillion has joined #openstack-ironic | 23:16 | |
| harlowja | NobodyCam devananda http://i.imgur.com/q8Hknb9.png (Created from that review) | 23:19 |
| harlowja | using http://paste.ubuntu.com/9448172/ (A clone of something similar in taskflow) | 23:19 |
| harlowja | useful for looking at (and seeing if any anomalies) | 23:19 |
| harlowja | rloo ^ | 23:19 |
| harlowja | feel free to hack that up, but neat that it can be done imho | 23:22 |
| *** mikedillion has quit IRC | 23:23 | |
| NobodyCam | harlowja: wow thats hard to follow ... but really cool | 23:23 |
| harlowja | NobodyCam hey, its from the FSM; can only blame the FSM :-P | 23:23 |
| harlowja | only outputs what u give it, ha | 23:23 |
| NobodyCam | lol :) | 23:23 |
| *** mikedillion has joined #openstack-ironic | 23:23 | |
| harlowja | i think u can add edge names to, the code above just isn't, | 23:23 |
| harlowja | that might help | 23:23 |
| NobodyCam | although rebuild looks wrong there | 23:25 |
| harlowja | ya, thats what the graph is nice for finding out | 23:25 |
| NobodyCam | :) | 23:25 |
| devananda | PaulCzar: this statement doesn't jive for me. what do you mean? "the pxe_ssh deploy images don't seem to get dns info" | 23:26 |
| devananda | NobodyCam: hm. ya. "rebuild" is an input, not a state, in the current machine | 23:27 |
| NobodyCam | :) | 23:28 |
| PaulCzar | devananda: bad wording? basically when its doing the pxe / bootloader thing it boots up and tries to curl the ironic-api ... it fails saying that it can't look up the name | 23:28 |
| devananda | PaulCzar: oh! shouldn't it be using the IP ? | 23:28 |
| harlowja | * http://i.imgur.com/SnjnZsF.png has the 'on event' names | 23:28 |
| PaulCzar | devananda: that would mean that keystone would need to have it as IP in the catalog | 23:28 |
| harlowja | for edges ^ | 23:28 |
| NobodyCam | harlowja: Much nicer | 23:28 |
| harlowja | * http://paste.ubuntu.com/9448289/ | 23:28 |
| PaulCzar | devananda: which would among other things break SSL certs | 23:29 |
| *** krtaylor has joined #openstack-ironic | 23:29 | |
| NobodyCam | harlowja: Neat-oh TY | 23:29 |
| harlowja | np | 23:29 |
| PaulCzar | devananda: if there was a way to tell ironic to override the endpoint and pass the IP to the bootloader instead then I'd be okay using the IP, because there's nothing scary in the payload | 23:30 |
| devananda | PaulCzar: the ironic API URL is passed as a kernel cmdline, not fetched from keystone: https://github.com/openstack/diskimage-builder/blob/master/elements/deploy-ironic/init.d/80-deploy-ironic#L1 | 23:30 |
| PaulCzar | devananda: right, but ironic sets that ... not me | 23:30 |
| PaulCzar | unless I'm being obtuse and there's an obvious place to override it | 23:30 |
| devananda | https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/pxe_config.template | 23:31 |
| devananda | https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/pxe.py#L75 | 23:31 |
| devananda | PaulCzar: config option :) | 23:31 |
| NobodyCam | PaulCzar: https://github.com/openstack/ironic/blob/master/etc/ironic/ironic.conf.sample#L520 (config option) | 23:32 |
| devananda | oops. NobodyCam is correct | 23:33 |
| devananda | I pasted a link to the wrong option | 23:33 |
| PaulCzar | devananda: NobodyCam ahahahahaha dammit. I had scoured the config looking for stuff like ironic_api ... | 23:34 |
| NobodyCam | :) | 23:34 |
| harlowja | zer0c00l u looked at any of the state machine stuff, might be interseting from a newb persepctive to see what u think | 23:35 |
| zer0c00l | harlowja: link? | 23:37 |
| harlowja | zer0c00l https://review.openstack.org/#/c/139215/ visualized @ http://i.imgur.com/SnjnZsF.png | 23:38 |
| harlowja | maybe requires to much context though | 23:39 |
| *** shakamunyi has quit IRC | 23:39 | |
| NobodyCam | brb | 23:40 |
| *** jb11211 has joined #openstack-ironic | 23:44 | |
| jb11211 | Hey guys. is there anything ironic which could snapshot hosts? say if you were sharing the same hardware and wanted to grab/restore images | 23:44 |
| *** shakamunyi has joined #openstack-ironic | 23:45 | |
| *** anderbubble has quit IRC | 23:46 | |
| *** jrist is now known as knowncitizen | 23:48 | |
| *** knowncitizen is now known as jrist_ | 23:49 | |
| *** naohirot has joined #openstack-ironic | 23:50 | |
| *** jlk has joined #openstack-ironic | 23:50 | |
| naohirot | good morning ironic | 23:50 |
| *** yuanying_ has joined #openstack-ironic | 23:50 | |
| *** jrist_ is now known as jrist | 23:50 | |
| *** jrist is now known as jrist_ | 23:50 | |
| *** jrist_ is now known as jrist | 23:51 | |
| NobodyCam | morning naohirot | 23:52 |
| naohirot | NobodyCam: Hi | 23:52 |
| *** Hefeweizen has quit IRC | 23:52 | |
| devananda | jb11211: nope | 23:53 |
| *** yuanying has quit IRC | 23:53 | |
| *** shakamunyi has quit IRC | 23:54 | |
| naohirot | NobodyCam: quick question, http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-compute-service-to-use-the-bare-metal-service | 23:56 |
| naohirot | NobodyCam: /etc/nova/nova.conf, This configuration file must be modified on the Compute Service’s controller nodes and compute nodes. | 23:56 |
| *** harlowja has quit IRC | 23:56 | |
| *** igordcard has joined #openstack-ironic | 23:57 | |
| naohirot | NobodyCam: does that mean that OpenStack of a pair of controller and compute node have to be prepared for 2 set of controller and compute node at least to manage both virtual deploy and physical deploy? | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
